More info: That happened on Nexus 5 Android 6 with WebView:51.0.2704.81

I can't symbolize the microdump. The microdump itself seems broken or mangled.
We got an "ILL_ILLOPC" error. What does that mean?

Meanwhile Satya:
could webview TE reproduce this error?

alekyoo@ working on it. We will update soon.

Used Nexus 5x with 51.0.2704.81 and 52.0.2743.91 - Unable to repro 

Actually the microdump symbolized to this:
 1  libwebviewchromium.so!blink::Platform::initialize [Platform.cpp : 92 + 0x7]
     r4 = 0xa44a7fc0    r5 = 0x8d590b00    r6 = 0xa4403a9c    r7 = 0xb6d6bec0
     r8 = 0x8d590b00    r9 = 0x892e3400   r10 = 0x93ebd8c4    fp = 0x8d593084
     sp = 0x93ebd650    pc = 0xa2bcc1a9
    Found by: call frame info
 2  libwebviewchromium.so!blink::initialize [WebKit.cpp : 82 + 0x3]
     r4 = 0x8d590b00    r5 = 0x93ebd6ec    r6 = 0xa4403a9c    r7 = 0xb6d6bec0
     r8 = 0x8d590b00    r9 = 0x892e3400   r10 = 0x93ebd8c4    fp = 0x8d593084
     sp = 0x93ebd660    pc = 0xa30d4385
    Found by: call frame info
 3  libwebviewchromium.so!content::RenderThreadImpl::InitializeWebKit [render_thread_impl.cc : 1206 + 0x7]
     r4 = 0x8d593080    r5 = 0x93ebd6ec    r6 = 0xa4403a9c    r7 = 0xb6d6bec0
     r8 = 0x8d590b00    r9 = 0x892e3400   r10 = 0x93ebd8c4    fp = 0x8d593084
     sp = 0x93ebd668    pc = 0xa3b6f2c9
    Found by: call frame info
 4  libwebviewchromium.so!content::RenderThreadImpl::Init [render_thread_impl.cc : 638 + 0x7]
     r4 = 0x8d593080    r5 = 0x93ebd7bc    r6 = 0xa4403a9c    r7 = 0x9dbbf400
     r8 = 0x93ebd8c0    r9 = 0xb6d6bec0   r10 = 0x93ebd8c4    fp = 0x8d593084
     sp = 0x93ebd768    pc = 0xa3b6f9eb
    Found by: call frame info
 5  libwebviewchromium.so!content::RenderThreadImpl::RenderThreadImpl [render_thread_impl.cc : 602 + 0x5]
     r4 = 0x8d593080    r5 = 0x00000000    r6 = 0x8d593224    r7 = 0xa431556c
     r8 = 0x93ebd834    r9 = 0x93ebd858   r10 = 0x93ebd8c4    fp = 0x8d593084
     sp = 0x93ebd828    pc = 0xa3b7052d
    Found by: call frame info
 6  libwebviewchromium.so!content::RenderThreadImpl::RenderThreadImpl [render_thread_impl.cc : 603 + 0xd]
     r4 = 0x98639e94    r5 = 0x93ebd8c4    r6 = 0x93ebd8c0    r7 = 0x8d593080
     r8 = 0x98639e94    r9 = 0xbee0c304   r10 = 0x93dbf000    fp = 0xa3c309f9
     sp = 0x93ebd8a8    pc = 0xa3b70583
    Found by: call frame info
 7  libwebviewchromium.so!content::RenderThreadImpl::Create [render_thread_impl.cc : 577 + 0x3]
     r4 = 0x00000000    r5 = 0x93ebd8c4    r6 = 0x93ebd8c0    r7 = 0x8d593080
     r8 = 0x98639e94    r9 = 0xbee0c304   r10 = 0x93dbf000    fp = 0xa3c309f9
     sp = 0x93ebd8b8    pc = 0xa3b705b7
    Found by: call frame info
 8  libwebviewchromium.so!base::Thread::ThreadMain [thread.cc : 245 + 0x3]
     r4 = 0x98639e60    r5 = 0xb6d6bec0    r6 = 0x9a0bc3e0    r7 = 0x95de1600
     r8 = 0xb6fc0b84    r9 = 0xbee0c304   r10 = 0x93dbf000    fp = 0xa3c309f9
     sp = 0x93ebd8e0    pc = 0xa3c33141
    Found by: call frame info
 9  libwebviewchromium.so!ThreadFunc [platform_thread_posix.cc : 70 + 0x7]
     r4 = 0x93ebd930    r5 = 0x98639e60    r6 = 0x9ee09e50    r7 = 0x00000078
     r8 = 0xb6fc0b84    r9 = 0xbee0c304   r10 = 0x93dbf000    fp = 0xa3c309f9
     sp = 0x93ebd908    pc = 0xa3c30a2d
    Found by: call frame info

There are thousands of the same crash signature in our internal crash/ site.

It's hitting this:
    RELEASE_ASSERT(!s_initialized);
So blink is being initialized twice... I think we have a bug about it somewhere.

If we can repro this (even occasionally) this would be *immensely* useful, because we have no repro case for 514141. Please just try to repro this with M51, because the problem appears to have vanished or changed in M52 from our current crash data and I'm not sure why :)